debase 0.1.18__py3-none-any.whl → 0.1.19__py3-none-any.whl

debase/_version.py

@@ -1,3 +1,3 @@
 """Version information."""
 
-__version__ = "0.1.18"
+__version__ = "0.1.19"

debase/reaction_info_extractor.py

@@ -927,12 +927,77 @@ Ignore locations that contain data for other campaigns.
     # ------------------------------------------------------------------
     # 6.3 Extract metrics in batch
     # ------------------------------------------------------------------
+    
+    def _validate_location_exists(self, ref: str) -> bool:
+        """Verify that the referenced location actually exists in the document."""
+        # Search for the actual reference in the document
+        for page_num in range(len(self.doc)):
+            page = self.doc[page_num]
+            text = page.get_text()
+            
+            # Look for table references like "Table 1", "Table S1", etc.
+            if re.search(rf'\b{re.escape(ref)}\b', text, re.IGNORECASE):
+                return True
+        
+        return False
+
+    def _validate_context(self, snippet: str, enzyme_list: List[str], ref: str) -> bool:
+        """Validate that the context contains meaningful content for extraction."""
+        if not snippet or len(snippet.strip()) < 50:
+            LOGGER.warning("Insufficient context for extraction from %s - skipping", ref)
+            return False
+        
+        # Check if context actually mentions the enzymes we're looking for
+        enzyme_mentions = sum(1 for enzyme in enzyme_list if enzyme.lower() in snippet.lower())
+        if enzyme_mentions == 0:
+            LOGGER.warning("No enzyme mentions found in context for %s - skipping", ref)
+            return False
+        
+        # Check for performance-related keywords
+        performance_keywords = ['yield', 'selectivity', 'conversion', 'ee', 'er', 'ttn', 'ton', 'tof', '%', 'percent']
+        has_performance_data = any(keyword in snippet.lower() for keyword in performance_keywords)
+        
+        if not has_performance_data:
+            LOGGER.warning("No performance metrics found in context for %s - skipping", ref)
+            return False
+        
+        LOGGER.info("Context validated for %s: %d chars, %d enzyme mentions", ref, len(snippet), enzyme_mentions)
+        return True
+
+    def _validate_response(self, data: Dict, enzyme_list: List[str], ref: str) -> bool:
+        """Validate that the response contains meaningful data for the requested enzymes."""
+        if not data or not isinstance(data, dict):
+            LOGGER.warning("Invalid response format from %s - skipping", ref)
+            return False
+        
+        # Check if we got data for at least one enzyme
+        enzymes_with_data = 0
+        for enzyme in enzyme_list:
+            enzyme_data = data.get(enzyme, {})
+            if isinstance(enzyme_data, dict) and enzyme_data:
+                # Check if there's at least one non-null metric
+                metrics = ['yield', 'ttn', 'ton', 'selectivity', 'conversion', 'tof', 'activity']
+                has_metric = any(enzyme_data.get(metric) is not None for metric in metrics)
+                if has_metric:
+                    enzymes_with_data += 1
+        
+        if enzymes_with_data == 0:
+            LOGGER.warning("No valid metrics found in response from %s - skipping", ref)
+            return False
+        
+        LOGGER.info("Response validated for %s: %d enzymes with data", ref, enzymes_with_data)
+        return True
 
     def extract_metrics_batch(self, enzyme_list: List[str], ref: str) -> List[Dict[str, Any]]:
         """Extract performance metrics for multiple enzymes from the identified location in batch."""
         ref_lc = ref.lower()
         image_b64: Optional[str] = None
         
+        # First, validate that the location actually exists in the document
+        if not self._validate_location_exists(ref):
+            LOGGER.warning("Location %s not found in document - skipping", ref)
+            return []
+        
         # Add campaign context if available
         campaign_context = ""
         if self.campaign_filter:
@@ -953,6 +1018,10 @@ Ignore locations that contain data for other campaigns.
         else:
             snippet = self._page_with_reference(ref) or ""
 
+        # Validate context before sending to Gemini
+        if not image_b64 and not self._validate_context(snippet, enzyme_list, ref):
+            return []
+
         enzyme_names = "\n".join([f"- {enzyme}" for enzyme in enzyme_list])
         
         if image_b64:
@@ -977,6 +1046,10 @@ Ignore locations that contain data for other campaigns.
                 image_b64=image_b64
             )
             
+            # Validate response has meaningful data
+            if not self._validate_response(data, enzyme_list, ref):
+                return []
+            
             # Handle the response format - expecting a dict with enzyme names as keys
             results = []
             if isinstance(data, dict):

{debase-0.1.18.dist-info → debase-0.1.19.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: debase
-Version: 0.1.18
+Version: 0.1.19
 Summary: Enzyme lineage analysis and sequence extraction package
 Home-page: https://github.com/YuemingLong/DEBase
 Author: DEBase Team

{debase-0.1.18.dist-info → debase-0.1.19.dist-info}/RECORD

@@ -1,17 +1,17 @@
 debase/PIPELINE_FLOW.md,sha256=S4nQyZlX39-Bchw1gQWPK60sHiFpB1eWHqo5GR9oTY8,4741
 debase/__init__.py,sha256=YeKveGj_8fwuu5ozoK2mUU86so_FjiCwsvg1d_lYVZU,586
 debase/__main__.py,sha256=LbxYt2x9TG5Ced7LpzzX_8gkWyXeZSlVHzqHfqAiPwQ,160
-debase/_version.py,sha256=Qd1kKsssesKE5FvJnDdAuZsx_BrxTSJJyt68SK99D54,50
+debase/_version.py,sha256=VbYiJzmzValsIDmCyQWPabFFsmy_TQ_Qp35j2mo-UKc,50
 debase/build_db.py,sha256=bW574GxsL1BJtDwM19urLbciPcejLzfraXZPpzm09FQ,7167
 debase/cleanup_sequence.py,sha256=QyhUqvTBVFTGM7ebAHmP3tif3Jq-8hvoLApYwAJtpH4,32702
 debase/enzyme_lineage_extractor.py,sha256=xbNKkIMRCM2dYHsX24vWX1EsQINaGSWBj-iTX10B8Mw,117057
 debase/lineage_format.py,sha256=IS9ig-Uv7KxtI9enZKM6YgQ7sitqwOo4cdXbOy38J3s,34232
-debase/reaction_info_extractor.py,sha256=W9CS0puFTdhJ_T2Fpy931EgnjOCsHHjbtU6RdnzDlhw,113140
+debase/reaction_info_extractor.py,sha256=otj8D3MnrThhUR_xOCc3sSVIw8hrCKnB4OY6y6NnaWA,116674
 debase/substrate_scope_extractor.py,sha256=9XDF-DxOqB63AwaVceAMvg7BcjoTQXE_pG2c_seM_DA,100698
 debase/wrapper.py,sha256=V9bs8ZiyCpJHMM5VuN74kiKdkQRVU6vyvLKCrO1BUB8,20890
-debase-0.1.18.dist-info/licenses/LICENSE,sha256=5sk9_tcNmr1r2iMIUAiioBo7wo38u8BrPlO7f0seqgE,1075
-debase-0.1.18.dist-info/METADATA,sha256=XvSrveJ0Y40c53JYUfiveaQNJ3qoEkxaQ61n3_--1cQ,10790
-debase-0.1.18.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-debase-0.1.18.dist-info/entry_points.txt,sha256=hUcxA1b4xORu-HHBFTe9u2KTdbxPzt0dwz95_6JNe9M,48
-debase-0.1.18.dist-info/top_level.txt,sha256=2BUeq-4kmQr0Rhl06AnRzmmZNs8WzBRK9OcJehkcdk8,7
-debase-0.1.18.dist-info/RECORD,,
+debase-0.1.19.dist-info/licenses/LICENSE,sha256=5sk9_tcNmr1r2iMIUAiioBo7wo38u8BrPlO7f0seqgE,1075
+debase-0.1.19.dist-info/METADATA,sha256=i1dFEB8kPkfTt8q8hJpAAAkZA29T2kb1bzPFMjzPdJU,10790
+debase-0.1.19.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+debase-0.1.19.dist-info/entry_points.txt,sha256=hUcxA1b4xORu-HHBFTe9u2KTdbxPzt0dwz95_6JNe9M,48
+debase-0.1.19.dist-info/top_level.txt,sha256=2BUeq-4kmQr0Rhl06AnRzmmZNs8WzBRK9OcJehkcdk8,7
+debase-0.1.19.dist-info/RECORD,,